激励与转角信息未知条件下桥梁结构物理参数识别  

Identification on physical parameters of bridge structures with unknown excitation and rotational response

作　　者：黄海新 金晓辉 程寿山 HUANG Hai-xin;JIN Xiao-hui;CHENG Shou-shan(School of Civil and Transportation Engineering,Hebei University of Technology,Tianjin 300401,China;National Engineering Laboratory of Bridge Safety and Technology(Beijing),Research Institute of Highway Ministry of Transport,Beijing 100080,China)

机构地区：[1]河北工业大学土木与交通学院,天津300401 [2]交通运输部公路科学研究所,桥梁结构安全技术国家工程实验室(北京),北京100080

出　　处：《计算力学学报》2024年第5期948-953,共6页Chinese Journal of Computational Mechanics

基　　金：天津市交通运输科技发展计划(2021-29,2023-48);桥梁结构安全技术国家工程实验室开放课题(2019-GJKFKT,2021-GJKFKT)资助项目。

摘　　要：桥梁结构物理参数识别往往需要完整的结构响应和已知的外加激励信息,这在实践中通常难以实现。为实现转角信息与外加激励数值均未知情况下的桥梁结构物理参数在线识别,先采用静力凝聚方法消去转动自由度,据此将扩展卡尔曼滤波方法与最小二乘方法结合,提出一种改进的扩展卡尔曼滤波方法。其中,针对静力凝聚后观测方程呈现高度非线性,导致卡尔曼量测矩阵计算低效的问题,引入基于克朗尼克积的分解算法。以一座简支梁桥为例,在移动荷载下进行了仿真测试。结果表明,所提方法能够准确地识别结构物理参数和外加激励,且克朗尼克积引入后,识别精度与程序运行速度均有显著提升。The physical parameter identification of bridge structures often requires complete structural response and known extended excitation information,which is usually difficult to obtain in practice.In order to realize the online identification of the physical parameters of a bridge structure when the information of rotation information and the applied excitation is unknown,an improved extended Kalman filter method is proposed by combining the extended Kalman filter method with the least-squares method after eliminating the rotational degrees of freedom by means of the static coalescence method.The decomposition algorithm based on the Kronecker product is introduced for the problem that the observation equation after static coalescence presents a high degree of nonlinearity,which leads to inefficient calculation of the Kalman measurement matrix.A simply supported girder bridge is simulated under moving load as an example.The results show that the proposed method can accurately identify the physical parameters of the structure and the applied excitation,and the accuracy of the identification and the running speed of the program are significantly improved after the introduction of the Kronecker product.

关 键 词：物理参数识别 激励和转角信息未知 改进扩展卡尔曼滤波 静力凝聚法 量测矩阵分解 

分 类 号：O327[理学—一般力学与力学基础]